Book review: The Halo Effect

When I read Built to Last, In search of ex­cel­lent and Good to great I im­me­di­ately thought: What a bunch of BS!!. But again, I think that about most busi­ness books.

The con­tent of these books seemed par­tic­u­larly lu­di­crous to me. I found both log­i­cal in­con­sis­ten­cies and method­olog­i­cal flaws in their process. For ex­am­ple, pick­ing suc­cess­ful com­pa­nies af­ter the fact is anal­o­gous to pick­ing the win­ners of a lot­tery and claim that they have ex­cep­tional skills. Betting the farm on a sin­gle con­cept (i.e. Big Hairy Goal) is clearly the right thing when you are right, but most of­ten than not you are wrong. What about all the com­pa­nies that did that and failed? Also, Keeping to the Core worked re­ally well for Coca Cola, but re­ally badly for Kodak.

These are just ex­am­ples. These books are packed full of such rea­son­ably sound­ing ab­sur­di­ties. I thought that every­body would im­me­di­ately spot this. I thought that the mediocre sub­se­quent per­for­mance of the com­pa­nies orig­i­nally se­lected as ex­cel­lent would ob­vi­ously demon­strate the un­sound­ness of such con­clu­sion. I was very wrong. These tomes are hugely suc­cess­ful. People seemed to dis­agree with me. They re­ally like this stuff. Given that my work and my life were not overly im­pacted one way or the other, I left the mat­ter to rest.

And then came The Halo Effect. It was a plea­sure for me to read, as it con­firmed my own ideas (confirmation bias any­one?) and added many more.

The ti­tle is slightly mis­lead­ing. A bet­ter ti­tle would have been: Why the idea of find­ing com­mon traits among suc­cess­ful com­pa­nies is baloney. I know, not as catchy. The catchy halo ef­fect, men­tioned in the ti­tle, is just the biggest delu­sion that peo­ple have about the topic.

I found an ex­per­i­ment re­counted in the book to be very il­lu­mi­nat­ing. They di­vided a set of peo­ple in mul­ti­ple groups and gave each group a prob­lem to solve. They then told half of the groups that they did a very good job and the other half that they did a very poor job. The thing is, they did so at ran­dom. They then asked peo­ple to rate the dy­nam­ics of their group on sev­eral fac­tors. People that were told that they did a good job re­ported to have had won­der­ful dy­nam­ics: open dis­cus­sions, in­spired lead­ers, high cal­iber peo­ple and such. The op­po­site was true for the other groups.

It did­n’t mat­ter what the ac­tual dy­nam­ics had re­ally been: some groups were very con­fronta­tional, oth­ers were very am­i­ca­ble; some were very vo­cal, oth­ers were very quiet. In the end, peo­ple who were told that they did good re­called the ex­pe­ri­ence in pos­i­tive terms. People who were told that they did poorly, re­called the ex­pe­ri­ence in neg­a­tive terms. Remember, good and bad out­comes were as­signed ran­domly. Also, the re­sults have been con­firmed in sev­eral sim­i­lar stud­ies.

It is not a good strat­egy that dri­ves per­for­mance. It is per­ceived per­for­mance that makes peo­ple think that they had a good strat­egy. This is: The halo ef­fect. This sim­ple con­cept has dra­matic reper­cus­sions.

First of all, it in­val­i­dates the main process used in these stud­ies: if you use news­pa­pers ar­ti­cles and man­agers rec­ol­lec­tion to draw con­clu­sions, you are just pil­ing ha­los over ha­los. People would cer­tainly give all sort of pos­i­tive at­trib­utes to com­pa­nies that suc­ceeded. They do it ex­actly be­cause they suc­ceeded. And when these com­pa­nies fail, they sud­denly give all sort of neg­a­tive at­trib­utes to them. The book is full of ex­am­ples of com­pa­nies that were rated as ex­cel­lent when their stock price went up and ter­ri­ble when their stock price went down. Nothing changed in these com­pa­nies, they were still do­ing the same ex­act thing.

The book con­tains many other in­sights of why these ‘studies’ are deeply flawed. It calls them delu­sions. You are prob­a­bly fa­mil­iar with some of them if you have a sci­en­tific back­ground: for ex­am­ple, Delusion of Correlation and Causality, Delusion of Single Explanation (factors are not in­de­pen­dent), Delusion of Connecting the Winning Dots (ad hoc af­ter the fact se­lec­tion). Other delu­sions are spe­cific to the an­a­lyzed do­main: Delusion of Lasting Success, Delusion of Absolute Performance, Delusion of Organizational Physics.

In the end what should man­agers do? Instead of fol­low­ing these book­s’s pre­cepts, they should fo­cus on two things: strat­egy and ex­e­cu­tion. Both of these things are prob­a­bilis­tic en­deav­ors where the ab­solutely wrong mea­sur­ing stick is the re­sult that you ob­tain. I al­ways found this to be a very com­mon er­ror peo­ple fall into at what­ever level in an or­ga­ni­za­tion. The er­ror is judg­ing the sound­ness of a de­ci­sion by its re­sult. A de­ci­sion should be judged by the sound­ness of the process to get to the de­ci­sion and the in­for­ma­tion avail­able at the time the de­ci­sion was taken. This is the only thing that is con­sis­tent with the prob­a­bilis­tic na­ture of the world we live in. The book pro­files three lead­ers who em­body this prob­a­bilis­tic view.

In sum­mary, please go and read this book. There are many more things in it than I de­scribed in this (albeit long and bor­ing) post.

Tags

• BOOKS
• FINANCIAL

5 Comments

Comments

wintermute

2008-01-11T17:50:41Z

The process to get the de­ci­sion and data avail­able at the time of the de­ci­sion are of course of para­mount im­por­tance. I agree on that. What I do not agree is that we should not rate the re­sult of the de­ci­sion to judge it. After all if the process is sound but the de­ci­sion we take leads to an un­wanted re­sult then the share­hold­ers will not be happy about that. In other words, if I had to bet my money I would bet it on some­one that is lucky. Lucky as Warren Buffet. I would not bet it with some­body with a good process like Myron Scholes.

lucabol

2008-01-11T20:04:22Z

I dis­agree. You can bet on the lucky one just af­ter the fact, and at that point he might be­come un­lucky.
BTW: Buffett has a bet­ter process than Scholes be­cause it is more tied to re­al­ity (assuming that by Scholes you mean the ef­fi­cient mar­ket the­o­rists).

Dolly Photon

2008-01-25T04:36:18Z

“The er­ror is judg­ing the sound­ness of a de­ci­sion by its re­sult” - you can’t pos­si­bly mean that! Otherwise it would be im­pos­si­ble to learn from ex­pe­ri­ence. Where else does our knowl­edge of the the “soundness” of our processes and in­for­ma­tion come from?

lucabol

2008-01-25T11:52:30Z

Excellent point !!!
Experience is a se­ries of ex­per­i­ments with an ini­tial state, an ac­tion and an out­come. It is the data base that helps us to form our opin­ion that given an ini­tial state (i) and an ac­tion (a1) there is a cer­tain prob­a­bil­ity (p1) that I ob­tain a par­tic­u­lar out­come (o1).
The ob­served re­sult of a de­ci­sion is just a mod­i­fi­ca­tion of the prob­a­bil­ity as­so­ci­ated with that ini­tial-state/​ac­tion pair. It just changes your opin­ion of p1, of­ten not by a large amount given that it is just one in­stance.
In this men­tal frame­work, you don’t judge the sound­ness of a de­ci­sion by the re­sult, you just change the as­so­ci­ated prob­a­bil­ity when you ob­serve a cer­tain re­sult.
This is not how peo­ple think, be­cause our minds are evo­lu­tion­ary wired for cause/​ef­fects re­la­tion­ships (i.e. I see a tiger, I bet­ter run). Nonetheless I ar­gue it is a bet­ter frame­work to cope with the cur­rent world (given the ob­served scarcity of tigers :) ).

Weddings

2008-06-06T05:26:30Z

When I read " Built to Last ", " In search of ex­cel­lent " and " Good to great " I im­me­di­ately thought: "What a bunch of BS!!". But again, I think that about most busi­ness books. The con­tent of these books seemed